AI agent of sports nutrition technologist

Budget: 25000 USD Deadline: 30 days

Look, there's a nuance - it's too early to evaluate the entire product with a single number, so I would suggest the first stage for 220,000 UAH and 30 working days. This includes the architecture, the agent prototype, a rules database for 2-3 markets, verification of the finished composition, generation of the recipe version, and basic documentation for the technologist. The full industrial version with an expanded country database, change log, roles, sources, and admin panel will be considered separately after the design phase.

The Ingello team has created similar AI systems and corporate platforms. It's important not just to connect the model, but to create a regulatory database, verifiable sources, dosage validation, and a mode where a live technologist remains the final decision-maker. Otherwise, the agent will sound confident but will make mistakes where the cost of error is unpleasant =)

> https://business.ingello.com/fractal - similar in terms of AI agents, step-by-step logic, and decision verification
> https://business.ingello.com/vorfahr - close in terms of automating complex domains and product logic
> https://business.ingello.com/lita - indirectly close in terms of medical IT and careful data handling

I would like to clarify two things - which markets are needed in the first version - the entire European Union, the USA, Canada, Britain - and which sources of norms should be prioritized?
Is it necessary to keep a history of recipe versions, the roles of the technologist and manager, and export the label to PDF, or is a report in the cabinet sufficient?

The main page of Ingello Systems for the exchange - https://systems-fl.ingello.com Expand

Budget: 19000 USD Deadline: 10 days

Hello. I will create a software agent that generates recipes for sports nutrition based on the specified product type and country, checks the composition for compliance with local regulations, and recalculates dosages when entering a new market. I will connect the regulations by country as a knowledge base so that the same product is correctly adapted for the EU and the USA, taking into account prohibited and restricted ingredients. I will start with a working MVP in two markets for you to evaluate the quality, and then we will expand the list of countries and products. The estimated cost is 19,000 rubles and 10 working days for the MVP; I will provide an exact estimate after clarifying the list of countries and sources of regulations. I am ready to start immediately.

Budget: 5000 USD Deadline: 35 days

Hello! My name is Oleg, I am the project manager for automation at Business Atlas. Your task is a cool large-scale case at the intersection of FoodTech, AI, and international compliance. We have extensive experience in building complex expert systems with knowledge bases for the markets of Ukraine, the USA, and Europe (over 50 successful projects implemented).
We build solution architectures based on low-code tools (Make, n8n) and CRM, ensuring flexibility, high speed of launch, and stability without overpaying for heavy custom code.
Here is the solution we are implementing for your terms of reference:
• AI Technology with a knowledge base (RAG system): We will connect Claude/GPT through a workflow in Make/n8n to a structured database of legislation from various countries (FDA, EU regulations, etc.). The AI will automatically scan dosage limits and permitted substance lists during the generation of recipes "from scratch" or checking the finished composition.
• Adaptation to new markets: The system will be able to take an existing recipe (for example, for the USA) with one click, compare it with EU standards, highlight prohibited components or dosage exceedances, and suggest safe alternatives based on scientific data.
• Documentation and label generation: The agent will automatically calculate nutritional values, generate correct text for labeling, and create basic technical specifications for production.
• Expert control (Human-in-the-loop): We design the system so that the AI produces a draft of the recipe project with detailed justification, which your live technologist can easily check, adjust, and approve in a single convenient interface (based on Airtable or a dashboard).
Message me privately — we will discuss which markets and types of products (proteins, pre-workouts) are currently a priority, and I will propose an architecture for a quick start!

Budget: 1500 USD Deadline: 7 days

Zlatoslava, a project considering EU/US regulations is precisely a task where logic, data, and verifiable rules are critical. I can help assemble the architecture of an AI agent: creating a recipe from scratch, checking the composition by country, adapting to the market, restrictions on ingredients and dosages, as well as forming data for labels and basic documentation. I have 7 years in web service and application development, I know how to build complex systems and work with a team. Let's discuss the details.

Budget: 1500 USD Deadline: 7 days

Good day, Zlatoslava!

Having studied the feedback from other specialists, I would like to offer my assistance with your project.

I will refrain from making suggestions for now, as it is important to understand the ultimate goals of creating your AI agent. To propose the optimal collaboration option, please clarify:

- Project launch timelines: is it a quick MVP creation or is there some time buffer?
- Is the technical specification ready or is there a vision for the future agent?
- What is the scope and categories of products that the agent should cover?
- Is development needed on a specific platform or can a solution be proposed?

The cost and timelines for development are influenced by:

1. Integration with legislative databases.
2. Complexity of algorithms for adapting recipes.
3. Availability or absence of a technical specification.
4. Interface requirements.
5. Support for various types of products.

At the initial stage, it is important to form a final vision of the AI agent. This will help understand what technologies and functions may be useful and how to improve the product. This approach will ensure compliance with legislative requirements and ease of use.

Your project requires deep knowledge in the field of sports nutrition and legislation. My experience in AI and web programming will help create a solution that takes all nuances into account.

The next step is to discuss the project in more detail. Let's fix the details in correspondence or at a meeting. Expand

Budget: 250 USD Deadline: 8 days

Hello. My name is Volodymyr.

I am an experienced developer with 8 years of experience in creating turnkey websites, mobile applications, and complex web systems.

I specialize in developing modern, responsive, and high-performance solutions. Over 8 years of practice, I have built a deep technical stack:

Frontend and interfaces: HTML, CSS, JavaScript, TypeScript, React.js, Next.js, Vite, Tailwind CSS, Bootstrap, and Elementor.

Mobile development: React Native and Flutter.

Backend and API: Node.js, PHP, Python, Laravel, REST API, and GraphQL.

Databases and caching: MySQL, MariaDB, PostgreSQL, MongoDB, and Redis.

Infrastructure and tools: Git, Docker, Nginx, Apache, Composer, and Linux (Terminal).

I effectively analyze client requirements, propose optimal architectural solutions, and strictly adhere to deadlines. In addition to development, I possess skills in design, technical SEO optimization, and project support. Expand

Budget: 600 USD Deadline: 8 days

Hello! I understood the task: an AI agent that generates and verifies sports nutrition recipes according to the regulations of a specific country (EU/US, etc.), with the final decision resting with your technologist.

First and foremost — this affects the entire accuracy: in a regulated field, the value is not in the agent's code (AI agents and RAG are our usual work), but in a structured database of regulations: ingredient → country → dosage limit → source (FDA/EFSA/WADA) → date of relevance. Regulations cannot be "hardcoded" into the model — it will become outdated and start inventing limits, which is unacceptable in this field. Therefore, the regulations live separately (DB), and the agent checks against it before each recipe and marks: approved / limited / prohibited — with justification for each ingredient.

How I will build it:
• LLM (Claude) + RAG over a structured database of regulations (PostgreSQL + pgvector).
• Generator agent — draft recipe based on product type and market; verification agent — runs the composition through the database, flags conflicts, calculates dosages based on scientific data.
• Output — recipe project + composition for the label + flags; the decision is made by your technologist (human in the loop — the right approach, I will lay it out this way).
• Start — MVP for 1-2 markets (EU/US), then we can scale to new countries using the same mechanism.

Key question, which affects the volume and exact price: do you provide the regulatory data (ingredients/limits by country, scientific dosages) — a ready source/structure — or is gathering and structuring them also part of the task? If you provide them — MVP will be faster; if it’s on us — that’s a separate stage, I will assess based on volume.

Cost: MVP (agent + RAG + verification against the provided database, 1-2 markets, human at the end) — $600, timeframe 8-10 days. Gathering/structuring regulations — as a separate stage based on volume.

Examples of our AI agents and automations: https://guardlabs.online/files/ai-portfolio.html Expand

Budget: 200 USD Deadline: 1 day

Hello!

We can create an AI solution for developing and testing recipes for different markets.

1. Which countries and product categories should we prioritize?
2. Is there a need to verify existing recipes or develop from scratch?

—
About us

We are dZENcode – a full-cycle IT development company: from design and programming to integrations and post-release support of digital solutions. We create projects from scratch and also connect to existing solutions that need refinement, development, or technical support.

You can find detailed information about our services and rates on our official website: Freelancehunt

Take a look – after that we can discuss the details and agree on the next step.

⚠️ After clarifying all the details, we will determine the scope of work, the suitable format of cooperation: task-based, outsourcing, or outstaffing – and the final cost.

With us, projects are guaranteed to reach release:
• 10+ years providing IT services;
• 90+ in-house specialists;
• 250+ public reviews since 2015;
• We support products under SLA after launch;
• We work under NDA and contracts with the company! Expand

Budget: 2900 USD Deadline: 21 days

Hello! I am creating LLM agents in Python for tasks where the response needs to be checked against a formalized rule base, rather than just "asking the model." The key here is not the agent itself, but where we get the regulations from: allowed ingredients and dosage limits by country need to be entered into a structured directory that the agent checks before each formulation—otherwise, there will be beautiful but incorrect compositions. Since the final decision is up to your technologist, I will display it directly in the interface: the agent provides a draft plus flags "limited / prohibited." It makes sense to start with an MVP for 1-2 markets (EU / USA), and then we can scale using the same mechanism. One question: do you provide the regulatory base by country, or is gathering that also my responsibility?

Budget: 4500 USD Deadline: 5 days

Good day! We will create such an agent turnkey — we specialize in LLM agents and bots for business, and we already have similar working solutions in our portfolio.

Here’s how I see it: the LLM-based agent takes the product type and target country and returns a draft formulation — with dosages, a list of prohibited or restricted ingredients, and the composition for the label. The logic of "different regulations — different composition" is embedded through a structured regulatory reference that the agent checks before responding. The final decision will remain with your specialist, which we will highlight directly in the interface.

Stack — Python + LLM, interface of your choice: Telegram bot or web chat. I recommend starting as an MVP for 1-2 markets (for example, the EU and the USA), and then we can gradually add new countries using the same mechanism.

One question regarding the specifics, to calculate accurately: do you provide the regulatory framework for the countries (reference documents, limits), or does the agent need to gather it from open sources? This greatly affects the scope of work.

Approximately MVP — 4-5 days. I am taking it at a starting price: I am gathering initial feedback on the platform, so I am making it advantageous. I can send live demo bots in the chat so you can immediately see the level.

Budget: 900 USD Deadline: 30 days

Good day, Zlata! I am interested in the project and will gladly take on its implementation.

Budget: 250 USD Deadline: 15 days

Hello! My name is Nikita. I have been implementing AI solutions in paid advertising and automating marketing processes for over 2 years, working with Google Ads, Meta Ads, and TikTok Ads.
✅What you get when working with me:
— An AI-enhanced advertising strategy instead of chaotic launches
— Automation of analytics and project economic control
— Systematic scaling based on data and AI tools
📈I work with projects of various scales and use AI for faster analysis of results, finding growth points, and optimizing advertising processes without unnecessary costs.
I am ready to discuss your tasks and offer a practical plan for implementing AI in your project's advertising.

Budget: 100 USD Deadline: 3 days

Hello, Zlata! I've built something like this before — if you just connect GPT without a database, it will produce a nice recipe but will make up the measurements. A separate database is needed where real standards for each country are stored — the agent checks each ingredient against it, and the technologist at the output sees not just the recipe, but immediately what is violated and why.

Here’s an example of the agent I built: huggingface.co/spaces/BlankD1/research-agent

$100 / 3 days — agent on LangGraph, standards for 3–5 markets, interface for testing.

Budget: 25000 USD Deadline: 45 days

I already have a practically ready similar solution for an AI agent with a knowledge base, rule checks, and document generation - it can be quickly adapted for sports nutrition and launch the first working prototype.

In terms of timelines, I would estimate 4-6 weeks for the first working version.
For the budget - starting from 180,000 UAH for a prototype with 2-3 markets, an ingredient database, restriction checks, recipe generation, and draft documents.
I will provide a more accurate estimate after clarifying the sources of the regulatory framework and the depth of calculations.

We can keep it simple at the start - build the core of the agent around a verifiable knowledge base, where each recommendation is linked to a source, country, dosage limit, and ingredient status.
Then we can add scenarios - create a recipe from scratch, check the composition, adapt it for a new market, prepare text for the label, and basic documentation.
A live technologist remains the final arbiter, while the agent provides a project solution and explains the logic - in this field, this is important because the cost of a mistake is higher than the cost of a pretty button =)

Questions regarding the task:
- Which markets are needed in the first version - EU, USA, Ukraine, UK, or others?
- Have the regulatory sources already been collected, or do they need to be gathered and structured on our side?

Relevant examples:
- https://business.ingello.com/vorfahr - AI and automation of complex expert logic
- https://business.ingello.com/fractal - agent processes and working with multi-step solutions
- https://business.ingello.com/lita - healthtech approach, where checks, roles, and responsibilities of the specialist are important.

Our freelance landing - https://systems-fl.ingello.com

In touch... we can discuss the first stage right here and quickly come to a clear launch plan. Expand

Budget: 160 USD Deadline: 2 days

Hello, Zlatoslava!

The task is clear — and the competitors are right about one thing: the main difficulty is not the agent itself, but the regulatory framework. If the agent starts to "infer" FDA or EFSA limits from general knowledge of the model, it will begin to provide incorrect dosages. This is unacceptable in a regulated field.

I work with the Claude API daily in production (aiscreener.best, @ai_cryptoanalyze_bot). Claude is the best at handling complex instructions and does not "hallucinate" outside the database.

Architecture:

1. Regulatory framework (PostgreSQL)
ingredient → country → dosage limit → source (FDA/EFSA/WADA) → date of relevance
This lives separately from the model — it is updated without rewriting code.

2. RAG + Claude
The agent checks the database before each recipe. The output is a draft with flags:
✅ approved / ⚠️ limited / ❌ prohibited — with justification for each ingredient.

3. Two agents
— Generator: creates a recipe from scratch for the product type and country
— Validator: checks the finished composition

4. Interface for the technologist
Web chat (Next.js) or Telegram — your choice.
Recipe history, export to PDF.

MVP: 2 markets (EU + USA), 14 days.
Then, using the same mechanism — new countries.

Key question: do you already have the regulatory framework (ingredients, limits by country) in some form — or is its collection and structuring also part of the task? Expand

Budget: 330 USD Deadline: 7 days

Hello! This is a very interesting project - I am happy to take on its implementation.

I have commercial experience in developing smart AI agents that work with specific knowledge bases and strictly regulated logic. In particular, I created AI solutions for a insulation company, where the agent had to clearly operate with the technical specifications of materials, building codes, and complex calculations for specific requests.

Budget: 530 USD Deadline: 14 days

It can be done on Next.js + backend (Node.js) with RAG architecture and a database of norms (EU/USA/other countries), where the agent will generate recipes, check ingredient restrictions, and adapt formulas for the market. But yes — the final validation still remains with the technologist, as specified in the technical specifications.

Budget: 60 USD Deadline: 3 days

Greetings! The task of creating an AI technologist is very interesting and relevant. I have experience working with Python and integrating AI models via API, I can set up system logic (System Prompts) and parse structured data.

I propose to implement an MVP version of this agent based on the combination of Python + OpenAI API / Anthropic API.

Here’s how I suggest doing it:

We will create a structured database of constraints (in JSON format or text instructions) for key markets (USA, EU).

We will set up prompt engineering so that the AI strictly follows dosage limits and provides formulations, dosage calculations, and basic text for labels without any "extra" information.

We will wrap this in a user-friendly interface (for example, a Telegram bot or a neat desktop application) so that the technologist can change parameters with a couple of clicks.

The cost includes basic setup of the agent's logic and testing on 2-3 test scenarios of products.

As soon as you approve my bid, I will share my username on Telegram — we can discuss in what format it would be most convenient for you to receive the finished project of the formulation (text, tables, or PDF report) and what initial data you already have. I would be happy to work on such an innovative task! Expand

Budget: 300 USD Deadline: 10 days

Hello, Zlatoslava! I am building exactly such AI agents for specific businesses — I have my own AI platform (Mercon), where the agent works strictly based on the database, without any fabrications.

For your task:
— a technology agent that selects recipes (proteins, bars, pre-workouts, isotonic drinks) based on your ingredients and restrictions;
— legislation from different countries — as a separate set of rules: the agent checks the recipe for compliance and highlights what is not allowed;
— answers with a link to the source (norm/ingredient), so the technologist can trust and verify.

The key is to ensure that the agent does not "hallucinate" about compositions and norms: this is solved through RAG + thoughtful prompt engineering, which I work on daily.

I suggest a short call — I will clarify the data sources for the norms and the format of the recipes.
Matviy Expand

Budget: 250 USD Deadline: 21 days

Hello! The project is very interesting — an AI agent for a sports nutrition technologist with a regulatory knowledge base requires thoughtful architecture, not just LLM integration.

I have 3.5 years of commercial experience in fullstack development (TypeScript, Vue/Nuxt, Next.js, NestJS, PostgreSQL). I have worked on products with RAG search, versioned knowledge bases, and structured data. I understand the importance of separating regulatory data from application code — so that updates to legislation do not require rebuilding the system.

I propose to implement: a RAG layer for searching regulatory documents, an ingredient database with dosage limits, an audit log of the agent's decisions, and an admin panel for managing data. I am ready to discuss the details and start soon.

Budget: 200 USD Deadline: 3 days

Interesting task, I have done similar things in conjunction with RAG and a database of regulatory documents. I see it like this: on the backend, there is a structured database of ingredients with their properties and permitted dosages by jurisdiction, the agent automatically validates the composition for the required country when creating the recipe and flags conflicts. Question: do you plan to upload legislative restrictions manually and update them yourself, or do you need integration with some source for updates? I am ready to discuss the details and approach to the MVP.

Budget: 500 USD Deadline: 10 days

Good day.
Our team has many years of experience in developing ERP, CRM, CMS, and specialized software for businesses. We create effective digital solutions that help automate processes, increase productivity, and scale companies.

We work with modern technologies — from bots and scripts to AI agents and analytical systems. We develop websites of varying complexity. In our portfolio, we have implemented ERP solutions for the hotel business, as well as for companies engaged in the import and sale of goods, and our own product XFitness — an ERP system specifically designed for fitness clubs.

We are ready to implement your project and offer the best solution tailored to your needs.
Our portfolio: Freelancehunt

We specialize in the following areas:
- Development of ERP Systems
- Development of CRM Systems
- Development of Websites of any complexity
- Development of CMS Systems
- Support for Websites
- Development of OpenCart
- Support for OpenCart
- Modification of OpenCart
- Refinement of OpenCart
- Development of WordPress
- Support for WordPress
- Modification of WordPress
- Refinement of WordPress
- Development of ECommerce
- Support for ECommerce
- Modification of ECommerce
- Refinement of ECommerce
- Development of Web Applications
- Support for 1C Servers
- Support for Web Servers
- Development of mobile applications
- Data parsing
- Development of bots
- Development of AI agents

and on the following technologies:
- Python
- PHP
- Laravel
- Symfony
- Yii2
- JS
- NodeJS
- jQuery
- TypeScript
- MySQL
- HTML
- CSS
- Vue
- Nuxt.js
- React
- React Native
- C++ Expand

Budget: 4000 USD Deadline: 28 days

Hello. The task is cool, but a regular LLM won't handle it out of the box — it will start hallucinating about dosages and laws, which is critical in the production of sports nutrition.

We need a hybrid: AI acts as a generator, while a robust backend on Laravel verifies everything against formulas. The regulations of countries (FDA, EFSA) will be stored in a vector storage pgvector directly within the PostgreSQL database.

At the code level, we will create two agents: the first drafts a recipe according to the specifications, and the second (compliance) automatically checks CAS numbers against the restrictions database and issues warnings if the limit is exceeded. A live technologist will manage the process and adjust compositions through a convenient admin panel on Filament v5.

To ensure you don't risk your budget, I suggest we proceed in phases:

1. MVP (2 weeks, $2000) — database architecture, integration with the model API (Claude/Gemini), logic for cyclical checks "technologist-lawyer" on the backend for one pilot country, and the admin panel. You will receive a ready working core for testing.
2. Production (another 2 weeks, $2000-2500) — automatic parsing of new laws from documents, scaling to any countries, and generating specifications/labels in PDF.

I have extensive experience in developing complex systems on Laravel, and I will set up the AI logic deterministically and without hallucinations.

In what format do you currently store the regulations and limits of countries (PDF files, databases, or links)? Are you ready to start with the MVP? Expand

Budget: 500 USD Deadline: 14 days

I have built similar solutions: AI workflow with conditional logic, dynamic context, and structured responses in a format convenient for verification by a specialist. The approach is as follows: the core of the agent is a system prompt with a database of regulations (EU/Regulation 1169/2011 + EFSA, USA/DSHEA + FDA, UA/DSTU), which is activated depending on the chosen market before each request. For each type of product (protein, pre-workout, isotonic, etc.), I prepare separate templates with mandatory fields: ingredients, dosage, market-specific prohibitive markers. The agent returns the formulation + list of risk positions + recommendations — in a format convenient for final verification by the technologist. Approximately 7–10 working days, ~$500–700 for MVP with 3 markets and 4–5 types of products. Is there already an existing database of regulations or a report format that your technologist uses — this will significantly affect the timelines?

Budget: 250 USD Deadline: 7 days

Good day. I am ready to take on the work. I will do it as you want. I am waiting for your messages, we will discuss the details.

Budget: 750 USD Deadline: 10 days

Hello! The architecture is clear: the LLM should not "know" the regulations from memory; they live in a separate structured database (ingredient -> country -> limit -> source -> date). The agent checks this before each formulation and marks the status of each ingredient. The final decision is up to the technologist; I will set it up correctly. Stack: NestJS + Claude API + PostgreSQL + pgvector (RAG for regulatory documents) + Vue 3 for the interface. One key question that affects the actual workload: do you already have the regulatory database (FDA, EFSA, ingredient limits) in a structured format, or is gathering it also part of the task? I propose an MVP for 2 markets (EU + USA), 3-4 product categories, with expansion using the same mechanism. Thank you for your response.

Budget: 2000 USD Deadline: 30 days

👋 Hello! The largest and most interesting projects— Freelancehunt

The task is very interesting! I am ready to develop an intelligent agent for creating sports nutrition recipes considering the legislation of different countries. The agent will be able to:

create recipes from scratch
check the composition for compliance with EU, US, and other market standards
adapt recipes for different countries
indicate restrictions and prohibited ingredients
generate data for labels

💶 Cost: up to $2000
⏳ Deadline: 14–30 days

I will create a convenient and useful assistant that will significantly speed up the work of the technologist. I am ready to discuss the requirements in detail and get started. How do you like this proposal? Expand

Budget: 200 USD Deadline: 5 days

Good day! I develop in Python, have worked on similar projects with React/Node.js, and am ready for collaboration.

Budget: 1200 USD Deadline: 21 days

Hello.
The project is interesting, I am ready to take on the MVP without undercutting and without promises that the model will know the legislation by heart. In such a task, this is the main risk.
I would build the agent not as a regular chat, but as a system with a separate database of regulations and ingredients:
ingredient;
country/market;
status: allowed, restricted, prohibited;
dosage limit;
product category;
source of the regulation;
date of relevance;
comment for the technologist.
On top of this, an AI agent works: it generates a draft recipe, checks the finished composition, adapts the formula for another market, shows conflicts, and forms data for the label. The final decision remains with the technologist, and that is correct.
How I see the MVP:
1. Structured database of regulations
For the first stage, it is better to take 1-2 markets, for example, EU/US, and 2-3 product categories. The regulations are not embedded in the prompt but stored separately so they can be updated.
2. Recipe agent
The user specifies the type of product, market, nutrient goals, and restrictions. The agent suggests a draft composition and explains why it chose the ingredients.
3. Verification agent
Checks each ingredient and dosage against the database: what is allowed, what is restricted, what is prohibited, where manual verification is needed.
4. Output for the technologist
Recipe, warnings, composition for the label, sources, list of contentious points.
Stack: Python, FastAPI, PostgreSQL/pgvector or Qdrant, OpenAI/Claude API, simple web interface for MVP.
From experience: I have done AI/RAG scenarios, backend systems with databases, workflows with manual verification, AI agents within business processes, parser/monitoring pipelines, and systems where it is important not to let the model fantasize freely. I can show GitHub and examples in correspondence/call.
If a full system for many countries and product categories is needed instead of an MVP version, it is better to consider that separately after discovery. Expand

Budget: 250 USD Deadline: 4 days

Good day, I have been in web programming for over 9 years. I work with REST APIs, frameworks, and CMS such as Django, Laravel, Yii2, WordPress, OpenCart, CodeIgniter, etc. I am ready to complete the task. Reviews: Freelancehunt

Budget: 1500 USD Deadline: 14 days

My team and I offer a full cycle of turnkey development. Our team consists of experienced developers, designers, and UX/UI specialists, which allows us to create a user-friendly and functional product that meets all your requirements. We will discuss timelines and pricing in private messages once we have a complete understanding of the scope of work. I look forward to your feedback. I can send examples of our work in private messages.

Budget: 100 USD Deadline: 5 days

I can develop it, but in any case, details will be needed for clarifications, thank you. I have experience in various projects.

Budget: 150 USD Deadline: 5 days

Hello!
I see that you need not just a chatbot, but an AI technology that generates sports nutrition recipes considering the regulatory requirements of different countries and automatically adapts them to specific markets.

In similar AI projects, I have built systems where LLM agents operated under a clear logic of data collection, verification, and processing through ChatGPT, Claude, Voiceflow, and Make.com. For this task, I would suggest creating a multi-level agent: a separate module for working with the database of legislative norms, a separate module for recipe generation, and a validation module that will check the composition for compliance with the country's requirements before issuing the result.

It is also possible to implement automatic documentation generation, a list of ingredients for the label, and explanations regarding prohibited or restricted components. It is especially important to implement a system for regular updates of the regulatory framework so that the agent does not operate on outdated data.

From my experience in integrating AI solutions, the biggest risk in such projects is not the generation of recipes, but the reliability of sources and control of the verification logic. That is why I would include a separate stage of automatic verification before passing the result to the technologist.

Could you please let me know: is the regulatory framework already compiled on your end (FDA, EFSA, and other regulators), or does it also need to be developed as part of the project?

Budget: 245 USD Deadline: 14 days

I understood the technical specifications: the AI agent creates recipes for sports nutrition products considering the legislation of the target country. The same protein bar for the EU and the USA may have different compositions because the permitted ingredients and dosage limits differ. The agent provides a draft recipe, which is then verified by a live technologist.

In terms of architecture, I see it like this: the core is an LLM agent (Claude or GPT-4o) with tool use for three external layers. The first layer is a RAG index based on regulations (FDA for the USA, EFSA and EU Novel Foods Regulation for the EU, Codex Alimentarius for the base) broken down by countries and product categories. Under it, there’s Qdrant or pgvector with embeddings from OpenAI or Voyage. The second layer is a tabular reference of ingredients with their categories, dosage limits, and notes on prohibitions by country (this is no longer RAG but structured SQL/JSON). The third layer is a function for calculating nutritional value and cost per 100g of the product.

When a user says to create a bar with 25g of protein for the EU market, the agent first loads the context of permitted ingredients for the EU from the second layer, then selects a combination that provides the desired profile through a simple constraint solver, checks each ingredient against the current regulations in the first layer, and outputs the composition plus citations from regulatory documents with a disclaimer that the final decision rests with the person.

One factor that will affect price and quality is the size of the regulatory database. If we are only talking about an MVP for the EU and the USA, this is a manageable database of 1-2 thousand documents. If we plan to cover about 10+ countries, this becomes a separate task for maintaining the database's relevance (regulations change, and the agent must see this). For the MVP, I recommend starting with 1-2 markets for validation, then scaling up.

In terms of timelines, realistically 14 working days for the MVP with one market (for example, the EU) and 3-4 product categories (bar, protein, isotonic, pre-workout). The stack is Python plus Claude or GPT-4o plus Qdrant plus a minimal web UI on Streamlit for testing. After validation, we can expand to the second market.

In my experience: AI agents with tool use and RAG have been my main focus for the past year. In production, I built a voice AI assistant for veterinary clinics on the same stack (RAG plus LLM as orchestrator plus tools), and an AI platform for contextual advertising with semantic search. I am ready to call to discuss the details. Expand

Budget: 500 USD Deadline: 7 days

Good day. We have 4 years of experience in web development and can implement an AI agent as a web service with a user-friendly interface for creating, verifying, and adapting sports nutrition recipes for different markets. The agent will be able to consider the legal requirements of individual countries, analyze ingredient and dosage restrictions, generate recipes, label composition, and basic documentation. For implementation, we propose an AI + RAG stack with a separate database of regulatory documents and the possibility of further expansion for new countries and product categories. Examples of our work: apple-family.com.ua/uk, 3magency.co.

Budget: 4000 USD Deadline: 30 days

Hello!

The key feature of your agent is "one product, different composition for the market" — which is also the main technical risk: if the model "remembers" the norms from memory, it will sometimes confidently make mistakes, and in additives, this is a legal issue. This is resolved by having the agent reason not from memory, but based on a verifiable database of regulations by country, with references to the source for each limit — so your specialist can confirm quickly, rather than rechecking from scratch.

To avoid being unsubstantiated: I can create a small demo on one product (for example, pre-workout) and a couple of markets in the EU/USA — I will show how the agent provides two different compositions and marks what is restricted and why.

The amount and timeline are a guideline for the MVP (one product + a couple of markets). I will provide an exact estimate after a short discovery phase, when we determine the markets and the source of regulatory data.

Message me privately about which product and market are priorities for you — I will prepare a demo for it.

Budget: 1000 USD Deadline: 30 days

Creating an AI agent for sports nutrition formulations is a task where most solutions fail not due to a lack of algorithms, but because of a superficial understanding of the regulatory frameworks of different countries.

It is important for the AI to not only consider prohibited ingredients but also to interpret legal nuances — for example, limits in the EU act as maximums, while in the US they are sometimes applied as minimums. This requires not template solutions, but a deep integration of legal sources into the generation logic.

The development of the agent will require mapping 30+ countries within 60 days, with a focus on dynamic adaptation of formulations. It will be necessary to determine what data format you want to use for input information — structured tables or free text from certificates.

Budget: 333 USD Deadline: 3 days

Hello! I will complete your task quickly and efficiently. I will create an agent.

My recent works
https://indexfast.pp.ua - fast website indexing
https://mono-bank.pp.ua - everything about Monobank
https://mamamia.pp.ua - online store
https://programist.pp.ua/ua/portfolio/ - portfolio of works
https://monitortest.pp.ua - monitor testing
https://keytest.pp.ua - keyboard testing
https://pctest.pp.ua - computer testing
https://vseetut.matviy.pp.ua - layout

My portfolio: https://freelancehunt.com/ua/freelancer/romas6ka.html#portfolio
Message me, I will start working today. I will be glad to collaborate with you! Expand

Budget: 200 USD Deadline: 21 days

Good day! I can implement an AI agent with a connection to an external knowledge base and regulatory data. This will allow updating requirements by country without modifying the logic and reduce the risk of errors from the AI.

Please send the details regarding the countries, sources of regulatory data, and the desired interface for the agent's operation.

Budget: 200 USD Deadline: 1 day

Hello! This is exactly the type of AI application where domain knowledge, regulatory data, and workflow design are much more important than just connecting an LLM to a chat interface. I understand that the goal is not to replace formulation development specialists, but to speed up their work by generating draft formulations that meet requirements, which are then reviewed and approved by an expert. The architecture I recommend combines an LLM layer with a structured database of regulatory documents and ingredients, rather than relying solely on the model's knowledge. The agent will be able to generate product formulations from scratch based on the product type and target country, check existing formulations for compliance with the regulatory requirements of a specific country, adapt formulations for new markets, identify prohibited or restricted ingredients, calculate ingredient dosages according to customizable scientific references, and generate draft label compositions and accompanying documentation. For reliable support of this, I would build a system based on a knowledge base of regulatory documents by country, a database of ingredients with restrictions and dosage limits, a RAG architecture for searching regulatory documents, an audit log showing why recommendations were made, and an admin panel for updating regulatory documents and ingredient data. A key requirement for such a project is that regulatory information remains editable and versioned, rather than hardcoded into prompts or application code. This allows for legislative updates without rebuilding the system. My preferred tech stack: Next.js, TypeScript, Python or NestJS, PostgreSQL, OpenAI or anthropological models, vector database for regulatory document search, and an admin panel for managing regulatory documents. I would also implement credibility indicators, source references, and transparent reasoning paths so that specialists can verify recommendations rather than viewing outputs as a "black box." This is a vivid example of AI application, as the system combines structured regulatory rules, scientific data, and expert assessment, rather than relying on autonomous decision-making. I would be happy to discuss the architecture stages and implementation details.

Budget: 200 USD Deadline: 4 days

Hello!

AI agents on RAG are my main topic, so the task is clear and interesting. I will describe how I see it, so it’s evident that I’ve understood it, not just skimmed through.

I am building the agent as a combination of LLM + knowledge base (RAG): it creates a recipe based on the product type and country, checks the finished composition for compliance, adapts it for the new market, flags prohibited/limited ingredients, calculates dosages, and forms the composition for the label. The fact that a live specialist checks the final output is the right approach: the agent provides a competent draft, and the person makes the decision. That’s how I will lay out the logic.

Let’s be honest about the main point right away, because the entire accuracy depends on this: the agent is only as good as the knowledge base underneath it — permitted/prohibited ingredients by country, dosage limits, scientific data. This is a regulated field, and if the agent starts to "infer" norms from the model’s general knowledge, it will begin to produce incorrect limits. Therefore, the core of the project is not so much the agent's code (that’s our usual work), but a reliable structured regulatory database that it relies on.

A key question that will determine the scope and timeline: do you provide the regulatory data (ingredients and limits by country, scientific dosages) / is there a ready source or structured database — or is gathering and structuring them also part of the task?

Your answer will significantly affect the architecture and assessment, so I want to understand this first and foremost. If the database exists — I will build the agent on top of it; if it needs to be collected — we will set that as a separate stage.

From which market and product type do you plan to start? It’s convenient to create an MVP in one country-one type, check the quality with your specialist, and then expand. Expand

Budget: 700 USD Deadline: 20 days

The key challenge is not the LLM itself, but the storage and updating of the regulatory database. EFSA, FDA, WADA — each has its own format, and the rules change constantly. It is not possible to embed them in the model: they become outdated quickly. Therefore, the legislation lives separately in PostgreSQL: ingredient → country → restriction → source → validity date. Synchronization is scheduled from the regulatory API.

On top of that, there are two agents. The first generates a recipe upon request (product type, target market, nutritional goal). The second runs each ingredient through the database and marks conflicts: what is approved, what is restricted by dosage, what is banned in a specific country. The output is a draft recipe with justification for each ingredient, with the final decision made by the technologist.

Interface. Hybrid: a structured form sets parameters (product type, target countries, nutritional goals), and the agent responds in the chat with a breakdown for each ingredient. There are two deployment options:

Telegram Mini App — the form opens in Telegram, and the result comes in the chat. Quick launch, no separate authorization required.
Web application (Next.js) — recipe history, version comparison, export to PDF.
Stack: Claude API (Sonnet) + LangGraph + PostgreSQL + pgvector, front-end — Telegram Mini App or Next.js.

Timeline and cost: MVP with Telegram interface, EU + USA — 2 weeks, $550–700. Web application with history and export — 3 weeks, $800–1000. Expand

Budget: 500 USD Deadline: 5 days

Good day!

My name is Ivan, I am the founder of Tyap Lyap AI.

We can develop an AI agent to assist a sports nutrition technologist: generating drafts of recipes, checking ingredients against the restrictions of the chosen country, adapting the composition for the EU/US market, calculating dosages, and preparing data for labels/documentation.

Timeline estimate: 5 days for MVP.

I am ready to discuss the details and propose a structure for implementation.

Budget: 500 USD Deadline: 14 days

Hello!

I have extensive experience in developing AI agents, RAG systems, and building expert assistants for automating business processes and working with documentation.

Implementation plan: to create an AI agent based on RAG with a database of regulatory requirements by country, a module for checking ingredients and dosages, generating recipes, adapting to EU/US markets, and forming documentation for technologists.

I suggest discussing the details of the architecture, sources of regulatory data, target markets, and MVP functionality in private messages.

Budget: 701 USD Deadline: 7 days

We have already communicated with you about another project, I can also implement this task, doing everything in the best way.)

Budget: 3250 USD Deadline: 25 days

Good day.

Very interesting project.

In my opinion, what is important here is not so much the LLM itself, but the correct architecture of knowledge and the system for checking compliance with regulations of different markets.

At RAI, we are engaged in the development of AI agents, RAG systems, and knowledge-base platforms, so we envision the implementation approximately as follows:

• a knowledge base on regulatory requirements of different countries (FDA, EFSA, and other regulators);
• a database of ingredients, permissible dosages, and restrictions;
• an AI agent for generating formulations;
• a module for checking compliance with the legislation of a specific market;
• a module for adapting formulations between countries;
• formulation of the composition for labeling and basic technical documentation;
• explanations of recommendations with references to sources and regulations.

For the MVP, I would recommend starting with 1-2 markets (for example, the EU and the USA) to quickly verify the quality of results and the correctness of the regulatory framework.

For a more accurate assessment, I would like to clarify:
• which countries are needed at the first stage;
• whether there is a proprietary database of regulatory documents or sources;
• whether a web interface is needed for technologists;
• whether automatic updates of the regulatory framework are planned.

I would be happy to discuss the details.

Best regards,
Tadeush Muratov
CEO, RAI (Robots Artificial Intelligence) Expand

Budget: 350 USD Deadline: 10 days

Hello, I am ready to take on this project, to complete it quickly and efficiently. We can discuss the details more in person.

Budget: 500 USD Deadline: 14 days

Hello! Great idea for an R&D product.

The main challenge in developing such an AI agent is to ensure it strictly adheres to regulations (FDA, EFSA) and does not "hallucinate" in the dosages of active substances. I know how to properly structure the knowledge base architecture so that the agent relies solely on actual regulatory documents and scientific data.

I can set up all the logic: from generating a draft formulation from scratch to an automatic checklist of prohibited ingredients when adapting the product for a new market (USA/EU).

Message me privately — we will discuss which LLM we will use for this implementation and where it’s best to parse the legislative database!

Budget: 300 USD Deadline: 30 days

Good afternoon! We have experience in developing AI agents for automating complex manufacturing processes. We implement solutions through the integration of LLM with current databases of international legislation, which will ensure the accuracy of formulations for different markets. We are ready to start on the system architecture and model training according to your standards.

Budget: 130 USD Deadline: 5 days

Hello, I have done similar projects but in a different theme, I have experience in implementing such projects.

I am currently available and can start working immediately, feel free to contact me.

Budget: 100 USD Deadline: 5 days

Hello.
The topic of sports nutrition is close to me; I periodically look at the compositions of pre-workouts and protein products and am constantly amazed at how different the formulas are for the USA and Europe.
I would build the solution so that the agent works through a separate database of regulations by country, checks the composition for restrictions, and explains, along with the formulation, why a particular ingredient was added, replaced, or excluded. This way, it will be easier for the specialist to conduct the final check.
Could you please let me know if the regulatory database by country has already been compiled or if it also needs to be developed as part of the project? For which markets is the first release needed — EU, USA, UK, or other countries? And is only a chat agent needed, or a full web interface for technologists?

Budget: 12345 USD Deadline: 3 days

Hello! Are you planning to launch the product simultaneously for several markets, such as the EU and the USA, or will you focus on one market first?
I will discuss the timelines and budget more precisely in personal correspondence.

Here’s how I will execute this project:
1. I will gather product requirements and identify target markets.
2. I will create a formulation considering the legislation of the chosen countries.
3. I will provide a finished composition for the label and basic documentation.

Thank you for considering my proposal. I look forward to the opportunity to collaborate with you!